The purpose of the proposed research is two-fold. First, it will explore the potential validity of using various types of health measures to explain labor market behavior as individuals near normal retirement age. Results from this work will then be incorporated into a structural model of retirement behavior. Data will come from the ongoing Health and Retirement Study (MRS). The upcoming availability of physical performance measures on the MRS provides an excellent opportunity to accomplish the first goal, and the Study contains precisely the variables needed to model the labor force behavior of older workers. The proposed structural model addresses the interplay between health, financial resources, and labor market behavior in the later part of working life. We propose to use well-specified longitudinal economic models to analyze how health status and economic factors jointly affect labor force behavior as workers approach retirement age, with particular focus on the choices people make when their health declines. Our work can be distinguished from the work being done by other economists in a number of important respects: first, much of our work will focus on the appropriate measurement of health; second, we will distinguish between various labor market choices older workers can make including job change (to accommodate health limitations) and applying for disability insurance. We intend the proposed research to serve several important functions. First, it should improve understanding of labor force behavior as people age, by tying together two important but hitherto largely separate lines of research: research on the effects of health and of economic factors, respectively, on retirement behavior. In doing so, it should provide timely evidence on the possible behavioral and fiscal effects of changes in such programs as Social Security, disability insurance, and Medicare. In addition, our findings should provide additional information on the quality and uses of the detailed health and economic data available in the MRS and help quantify the possible limitations of using less detailed datasets. Relevance for Public Health: The proposed research has the potential to influence, at a very fundamental level, the way that researchers measure health when studying behavior. It will also provide important evidence on the effects of various possible reforms of the Social Security System on the well being of those who near retirement age in poor health. [unreadable] [unreadable] [unreadable] [unreadable]